A simple model of the turnover of organic carbon in a soil profile: model test, parameter identification and sensitivity

Abstract. Simulation models are potentially useful tools to test our understanding of the processes involved in the turnover of soil organic carbon (SOC) and to evaluate the role of management practices in maintaining stocks of SOC. We describe here a simple model of SOC turnover at the soil profile scale that accounts for two key processes determining SOC persistence (i.e. microbial energy limitation and physical protection due to soil aggregation). We tested the model and evaluated the identifiability of key parameters using topsoil SOC contents measured in three treatments with contrasting organic matter inputs (i.e. fallow, mineral fertilized and cropped, with and without straw addition) in a long-term field trial. The estimated total input of organic matter (OM) in the treatment with straw added was roughly three times that of the treatment without straw addition, but only 12 % of the additional OM input remained in the soil after 54 years. By taking microbial energy limitation and enhanced physical protection of root residues into account, the model could explain the differences in C persistence among the three treatments, whilst also accurately matching the time-courses of SOC contents using the same set of model parameters. Models that do not explicitly consider microbial energy limitation and physical protection would need to adjust their parameter values (either decomposition rate constants or the retention coefficient) to match this data. We also performed a sensitivity analysis to identify the most influential parameters in the model determining soil profile stocks of OM at steady-state. Input distributions for soil and crop parameters in the model were defined for the agricultural production area of PO4 (east-central Sweden), which includes Uppsala. The resulting model predictions compared well with aggregated soil survey data for the PO4 region. This analysis showed that model parameters affecting SOC decomposition rates, including the rate constant for microbial-processed SOC and the parameters regulating physical protection and microbial energy limitation, are more sensitive than parameters determining OM inputs. Thus, the development of pedotransfer approaches to estimate SOC decomposition rates from soil properties would help to support predictive applications of the model at larger spatial scales.